d’identifier les raisons de biais éventuels dans ces simulations, notamment en ce
qui concerne le climat de la basse troposphère – une attention particulière sera
portée aux aspects touchant aux flux d’énergie près de la surface ainsi qu’aux
termes déterminant le bilan de masse en surface de la calotte de glace ;
d’effectuer des simulations complémentaires, avec des configurations modifiées
(résolution, choix des paramétrisations, scénarios d’émissions supplémentaires,
etc.).
Un aspect important concerne l’évaluation systématique de l’apport de la méthode choisie
de régionalisation. Pour ce faire, on effectuera une suite de simulations à haute et basse
résolution spatiale, avec et sans correction des biais climatiques des simulations globales
CMIP qui servent de « matériel de base » pour la régionalisation.
English version
Antarctic climate change : Analysis of the added value of variable resolution global
climate models
The ongoing climate change is rapid and global. The expected future change is also rapid
and global, but it will also show a large spatial heterogeneity. For example, it has been
known for a long time that warming will be particularly strong in the Arctic, and generally
stronger over the continents than over the oceans. This is also the spatial structure of the
observed ongoing warming. In order to evaluate the impacts of climate change, reliable
high-resolution information is therefore required, but this is not available using global
coupled climate models, the classic tools applied to produce large-scale climate
projections. Concerning the Antarctic climate change, one needs to be able to produce
climate change simulations at spatial scales of the order of a few tens of kilometers
because of the strong topographic gradients in the coastal regions (of the order of 2000 m
over typically some 100 km).
Global climate change projections are usually carried out in the framework of large-scale
international projects called CMIP. Recently, the scientific community got organized in
order to produce regional climate change simulations, linked to CMIP, in an international
effort called Coordinated Regional Downscaling Experiment (CORDEX). CORDEX has several
aims: To produce climate change projections usable for regional-scale climate impact
studies, and to make progress in the methods and techniques of regional climate modeling
by intercomparing model results.
This thesis takes place within Polar CORDEX (http://www.climate-
cryosphere.org/activities/targeted/polar-cordex). The aim is to produce Antarctic climate
change simulationsfor the end of the 21st century under a strong, unabated emission
scenario following approximately the current emission trajectory. We will use two French
atmospheric general circulation models (AGCM) which both have the stretched grid
capability, allowing for high-resolution climate change simulations over a region of
interest within a consistent global modeling approach. The detailed tasks of the PhD thesis
are:
To carry out climate change simulations with these models;
To evaluate these simulations by comparing them with available CORDEX
simulations carried out using regional limited-area climate models such as WRF,
RACMO and in particular MAR, and of course to observations;
To identify reasons for possible biases in these simulations, in particular
concerning lower tropospheric climate – particular attention will be paid to surface